vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported43 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [LoRA] Remove engine-side adapter on runtime unload
- [LMCache Connector] skip remote KV lookup when local prefix hit is sufficient
- [Attention] Add no-output path for custom backends
- Allow LMCacheConnectorV1 to support hybrid KV loads
- [Bugfix] Dedup gpu_block_ids in SimpleCPUOffload eager store path
- [RFC]: Attention Backend Refactor
- [Feature]: Interrupt Reasoning on Pattern
- [Bug]: LoRA on AWQ-quantized Llama-3.1-8B / Llama-3.2-3B produces degenerate output (same LoRA infra works on AWQ Mistral and on FP8 / BF16 Llama)
- [Bug]: FA2 paged-KV may read stale block-table tail entries
- [Bug]: FA2 partial-block clamp can load unwritten padded KV slots
- Docs
- Python not yet supported